Non‐Invasive Classification Approach for Spinocerebellar Ataxia Type 3 via Metabolic Fingerprints Enhanced by Bimetallic Alloys

ABSTRACT Spinocerebellar ataxia type 3 (SCA3) is a progressive inherited neurodegenerative disorder for which accessible blood‐based classification approaches are needed. In this retrospective, single‐center study, plasma metabolic fingerprints were acquired from 202 participants using mesoporous PdPt nanoparticle‐assisted laser desorption/ionization mass spectrometry (LDI‐MS) and analyzed using a Tabular Prior‐data Fitted Network (TabPFN) classifier. The model achieved AUCs of 0.952 in the discovery cohort and 0.963 in the internal temporal validation cohort. A reduced 21‐metabolite panel retained discriminatory performance, with AUCs of 0.901 and 0.932, respectively. Exploratory clinical and pathway analyses provided additional biological context. These findings highlight the potential of PdPt‐assisted plasma metabolic fingerprinting for SCA3 classification.